Files
Abstract
Building on previous accomplishments involves literature reviewing and knowledge synthesis. The challenge of reviewing is the overwhelmingly large volume of publications and the limited manual strategies and tools available to support the quick identification of salient publications. The process of knowledge synthesis is laborious because the technology for supporting this process is limited. To support the identified challenges, this dissertation proposes a set of methods based on digitizing core knowledge (causal or process models) as graphs. This digitized knowledge forms a knowledge network, and graph analytics, social network metrics, and natural language processing methods, among others, can be applied to accelerate knowledge exploration and synthesis.